Premium
A Bayesian nonparametric approach to causal inference on quantiles
Author(s) -
Xu Dandan,
Daniels Michael J.,
Winterstein Almut G.
Publication year - 2018
Publication title -
biometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 2.298
H-Index - 130
eISSN - 1541-0420
pISSN - 0006-341X
DOI - 10.1111/biom.12863
Subject(s) - quantile , causal inference , bayesian probability , econometrics , nonparametric statistics , inference , bayesian inference , frequentist inference , computer science , bayesian statistics , statistics , machine learning , mathematics , artificial intelligence
Summary We propose a Bayesian nonparametric approach (BNP) for causal inference on quantiles in the presence of many confounders. In particular, we define relevant causal quantities and specify BNP models to avoid bias from restrictive parametric assumptions. We first use Bayesian additive regression trees (BART) to model the propensity score and then construct the distribution of potential outcomes given the propensity score using a Dirichlet process mixture (DPM) of normals model. We thoroughly evaluate the operating characteristics of our approach and compare it to Bayesian and frequentist competitors. We use our approach to answer an important clinical question involving acute kidney injury using electronic health records.